The interaction of cells with the extracellular matrix can alter cell responses and is regulated by integrins on the cell surface. We used monoclonal antibodies to the VLA-4 integrins CD29 and CD49d followed by an F(ab')2 fragment of rabbit anti-mouse immunoglobulin G1 to crosslink integrins on the surface of human lung mast cells and basophils. Crosslinking either CD29 or CD49d caused a significant histamine release (HR) from the basophils of most asthmatic donors (10 of 14 for CD49d and 7 of 10 for CD29) (HR = 21 +/- 5%, n = 10, P < 0.005 for CD29 and HR = 19 +/- 4%, n = 14, P < 0.01 for CD49d) yet failed to initiate HR from the basophils of non-atopic and atopic donors (HR was 1 +/- 0.5% for CD29 and 1 +/- 0.5% for CD49d, n = 10, P = NS). Crosslinking either CD29 or CD49d also failed to initiate histamine release from human lung mast cells (HR was 1 +/- 1% for CD29 and 2 +/- 1% for CD49d). The basophils of asthmatic donors responded to 100 and 30 micrograms/ml tissue fibronectin (HR = 12 +/- 2% and 10 +/- 3% for 100 and 30 micrograms/ml fibronectin, respectively, n = 18, P < 0.05), whereas basophils of nonasthmatic patients again failed to degranulate (HR was 0 +/- 0.4% and 1 +/- 0.6%, respectively, n = 11, P = NS). In contrast to the basophil, crosslinking of either CD29 or CD49d failed to initiate histamine release in human lung mast cells (HR = 1 +/- 1% for CD29 and 2 +/- 1%, n = 15). Human lung mast cells were also unresponsive to tissue fibronectin (100 and 30 micrograms/ml) (HR = 1 +/- 1%, n = 5). The tyrosine kinase inhibitor, genistein, significantly reduced CD29- and CD49d-induced HR (inhibition = 83 +/- 7% for CD29 and 77 +/- 6% for CD49d, n > or = 5, P < 0.05). A second tyrosine kinase inhibitor, piceatannol, also significantly reduced both CD29- and CD49d-induced HR (inhibition was 62 +/- 19% for CD29 and 56 +/- 14% for CD49d, n = 7, P < or = 0.05). Integrin crosslinking also affected the response to a second, immunoglobulin E (IgE)-dependent stimulus. Both CD29 and CD49d clustering significantly inhibited anti-IgE-induced histamine release from the human basophil. Inhibition was 30 +/- 5%, n = 18, P < or = 0.001 for CD29 versus 40 +/- 6% for CD49d. In summary, we have shown that crosslinking the beta 1 integrins using either monoclonal antibodies or tissue fibronectin can initiate mediator release from the basophils of asthmatic patients by a mechanism which appears to be tyrosine kinase-mediated. In addition, clustering of integrins modulates the response to a second IgE-dependent signal.